by Curt Stager
By tapping into the professional grapevine, I soon found that scientists from the University of New Hampshire, Middlebury College, and the U.S. Geological Survey were faulting the NERA report for its reliance on data that yield “spurious temperature trends.” Applying more carefully vetted weather records to the same region, those scientist-critics found more widespread warming than the earlier report did, and some of the differences were quite striking. For example, NERA’s finding that most of Maine cooled off during the twentieth century conflicted with the newer data, which showed that more of it had, in fact, warmed along with the other New England states.
Although the new studies knocked bigger holes in NERA than Mike and I had, they didn’t leave us gloating. The dataset under attack was the same one that we had obtained from the National Climate Data Center, which meant that some of our own results were wrong, too.
The mistake was innocent enough, though regrettable. Rather than try to sift local climate trends out of a broader study conducted by others, we had gone directly to the raw weather measurements. But what we, the NERA team, and quite a few other investigators didn’t fully appreciate at the time was that this approach produces a host of more or less random errors in the magnitudes, and even the directions, of climatic trends.
Thus chastened and newly enlightened, I was led to the straight and narrow path of good clean data by Jerome Thaler, a climatologist from southern New York whose book Adirondack Weather caught my attention from the shelf of a local bookstore.
“The first thing you have to realize,” he explained over the phone, “is that these data are collected by real people, most of whom are unpaid volunteers.” In other words, the individual habits of people running a station can strongly affect the data that they collect.
For instance, Jane runs the local weather station for twenty years. She awakens early each morning to record the temperature before heading off to work—unless she’s on vacation or the kids are sick. That leaves a gap in the daily readings and distorts the monthly average temperature calculations.
When Jane retires, John offers to take over her duties. But John doesn’t like to get up early, so he takes his readings later in the morning when the sun starts to warm things up for the day. Automatically, and incorrectly, the daily temperature averages become warmer.
And then there are equipment upgrades, power outages, changes in the numbers of readings per day, changes in station location, and changes in local vegetation, all of which can affect temperature data. If those factors aren’t recorded carefully along with the weather measurements, then there’s no way to correct their influences on the true temperatures.
“When climatologists say that they’ve cleaned or adjusted a weather record, it’s not cheating, as some might claim,” Thaler continued. “You simply have to do it in order to correct the errors as accurately as you can.” The current best source of weather data, according to Thaler and every other expert on my contact list is the U.S. Historical Climatology Network (USHCN), the same source that was used to challenge NERA’s findings.
USHCN staff select only those stations that fully document such error sources. They correct for informational gaps as well as changes in physical settings and methodologies, and they explain their methods in deadly detail on a website that offers regularly updated data in chart or table form.
And what do the new and improved USHCN records tell us about the Adirondacks? Several local weather stations have still cooled slightly since the 1950s, but the trend at Wanakena, in the northwest corner of the park, no longer dropped as it did with our raw dataset. Averaging all of the Adirondack station curves into one bouncy line showed the ups and downs balancing out into a slight overall rise of mean annual temperature during the twentieth century, but an upswing since the early 1970s warmed the whole region more steeply, by about a degree and a half, thereby more closely echoing the global pattern. The change was most significant, in a statistical sense, during June and September. December temperatures also rose, but they did so too erratically for the trend to be distinguished from random variability, and the other months of the year showed no major trend one way or the other.
Okay, now I believe it. We’re warming up here in these mountains, too. But how fast? If you look only at the last fifty years, then you see a much weaker average warming than what you find if you examine the last thirty years. So how should we decide which time period to use?
Most experts define a region’s present climate with a sliding time window that averages the last thirty years of weather conditions. For example, a scientist speaking in 1990 would have described the climate of the Adirondacks by summarizing the previous thirty years of weather data spanning 1961 to 1990. Someone doing the same thing a decade later would have examined data from 1971 to 2000, and so on. Therefore, the last three decades are a reasonable time frame to focus on for a study of local climate. In addition, reputable studies show that the post-1970 global temperature rise is more clearly attributable to greenhouse gas buildups than to any other factor, such as changes in the sun’s output or clearing skies. By those criteria, we can justifiably treat a recent thirty-year average warming trend as a sign that global change is probably at work here like a slowly rising tide beneath the erratic chop of short-term weather fluctuations.
In 2006, a new team of investigators from several northeastern institutions published a report entitled “Northeast Climate Impacts Assessment” (NECIA). To my relief the regional weather histories in the report more or less resembled my own updated Adirondack reconstructions. So, too, did a subsequent analysis of upstate New York weather records that was conducted as a master’s thesis project by Kathie Dello at the State University of New York, Albany. NECIA’s work still made no Adirondack-specific predictions, but knowing that both local and large-scale climates really have begun to change in similar fashion during the last three decades made their conclusions and my own seem more reliable.
It took a lot of work to convince me that the Adirondacks are warming. Not because it seemed particularly unlikely, but simply because there was little solid evidence available to support that claim until recently. Furthermore, it can be difficult to link local changes conclusively to the global greenhouse effect because the large-scale trends that it produces are slow and gentle compared to the erratic flip-flops of short-term, small-scale weather, especially up here in these mountains. It’s a shame that those who deny the existence of global warming out of stubbornness or willful ignorance so often bear the epithet “climate skeptic” (rather than something like “naysayer” or “contrarian”), because the job of a good scientist is to maintain reasonable skepticism in the face of a story that’s long on sex appeal but short on facts. Until someone down-scales global climate change information to fit your particular piece of the planet with demonstrable accuracy, you can’t really be sure whether it moves in the climatic mainstream or not.
Here in the Adirondacks, though, the long-term warming trend that is now firmly documented by local weather records is already producing some clearly observable changes in the realm of winter ice. According to the NECIA report, ice-out is now coming earlier to lakes throughout the northeastern states than it did a hundred years ago, though that’s not necessarily temperature at work. Snow and wind also control the fate of lake and river ice. Melting happens later in spring if the ice is well insulated by a blanket of snow, and that blanket thins or thickens depending on how much snow falls, how much melts, and how much blows away. And what usually makes lake ice break up at any given moment is wind, not heat alone. In calm weather, lake ice can sit for days, rotting into brittle, vertically packed needles, as it waits for the first breeze to smash it against the downwind shore.
A clearer indicator of warming is the date of freeze-up, because it is unaffected by the complicating effects of snow cover and midwinter ice thickness. Moreover, our local warming trend in autumn far exceeds that in spring, and this seasonal imbalance is reflected in the behavior of our lake ice. Thaler
’s book presents a century-long record of ice cover from Mirror Lake, an attractive body of water in downtown Lake Placid, and local librarian Judith Shea helped me to update that record by contacting a boating club that runs an annual ice-out contest there. Those data show that Mirror Lake now freezes two weeks later than it did during the early 1900s; the more erratic ice-out record, on the other hand, yields only a weak trend toward slightly earlier dates. And at lower elevations, where temperatures are already warmer to begin with, the change is even more obvious. Lake Champlain, which occupies a long valley on the eastern border of the Adirondack Park, hasn’t frozen over at all in recent winters. Records dating back to the early 1800s show that Lake Champlain skipped its winter freeze-up only three times during the nineteenth century but more than two dozen times since 1950. That’s difficult to explain with anything but a warming trend.
I published these and similar results with several colleagues in a recent issue of the Adirondack Journal of Environmental Studies, and when the Associated Press later interviewed some of us about the regional retreat of lake ice, I was amused by the public response that the article triggered. Almost all the comments that I found posted online were negative. One writer called it “a cherry-picking promotion for the discredited global warming movement.” Another typical posting said, “Global warming can be traced primarily to the fevered ravings emitted by ecofreaks.” Having previously been accused by the environmental community of not being alarmist enough, I took these opposing comments to be a good sign for a scientist. When you’re bashed by both sides in an emotionally charged argument, I like to believe that it means you’re standing in the middle where the truth probably lies. But the humor of the situation quickly faded when, just a few weeks later, three people lost their lives by falling through the ice on a local lake, and the U.S. Coast Guard warned snowmobilers and fisherfolk that the ice on Lake Champlain has become treacherously thin.
Freeze-up dates on Lake Champlain since the early 19th century. National Weather Service Forecast Office, Burlington, VT
Another potentially useful source of information about the effects of climate change on specific locales such as the Adirondacks is less scientific in nature: the informal observations of amateur naturalists. A former student of mine, Brendan Wiltse, investigated the environmental history of Lake George for his senior thesis by comparing modern water quality conditions near his family’s lakeshore home to those of the past as revealed in sediment cores and historical documents. One of his key sources of information was Thomas Jefferson, who visited the area in 1791. Jefferson kept written accounts of his natural history observations during that visit, which show what the beautiful lake and surrounding forests were like two centuries ago. “Its water is limpid as crystal,” he wrote, “and the mountainsides are covered with rich groves of fir, pine, aspen, and birch down to the water’s edge.” Murkier conditions today in some parts of the lake therefore warn of a troubling decline in water quality due to recent human activity. Unfortunately, Jefferson’s visit didn’t last long enough for him to record any useful weather observations that we could also compare to present conditions.
Records of ecological change in more modern times can be helpful, too, thanks to both professional and amateur naturalists. Since 1991, I’ve been recording when the first robins arrive on the Paul Smith’s College campus, when the native solitary bees emerge from their burrows on the sunny, south-facing slope of Essex Hill, when the fiery red buds burst on the maples that grow next to Cantwell Hall, and when the wood frogs and the black-and-yellow-spotted salamanders migrate to their ephemeral meltwater breeding pools on Keese Mills Road. This has become something of a ritual for me over the years, a way to feel more closely connected to the other species living around me and to better sense the rhythm of the seasons. And, of course, it has also allowed me to watch for local signs of change, which is why I’m careful to make my observations in precisely the same locations every year.
Most of the creatures and plants on my watch list have made no statistically significant changes in when they show up, wake up, or open up since 1991, probably (as local weather data show) springtime temperatures haven’t risen significantly here in recent decades despite the warming at other times of year; perhaps it’s also because I haven’t been watching long enough yet to detect subtle underlying trends amid the random variability. While there’s little sign of change in my observational data, I wouldn’t even have known about that lack if I hadn’t kept watching for as long as I have.
But some North Country residents have been at this sort of thing for much longer than I have, as I discovered by putting the word out on our local radio station. For the last two decades I’ve cohosted Natural Selections, a weekly science program on North Country Public Radio with news director Martha Foley, and in May of 2007 we dedicated an hour-long call-in session to signs of climate change in the area. The response was both entertaining and informative.
A listener from Fort Covington called in to say that her elderly father has documented the return of redwing blackbirds to their favored cattail marshes every spring since 1969. According to his handwritten notes, the birds aren’t showing up any earlier or later, on average; again, this is probably because spring temperatures here still show little or no sign of a greenhouse-driven trend.
Biologist Stacy McNulty from the Huntington Wildlife Forest passed along some records of Trillium and hobblebush flowering dates that began in the mid-1970s, along with ice-cover data from local lakes. No significant trends among the flowers or ice-out dates yet, but freeze-up is happening significantly later on the lakes.
Jeff Chiarenzelli, a geologist at Saint Lawrence University, sent in a compilation of discharge records from several Adirondack rivers that showed a significant increase of runoff during the twentieth century, especially since the early 1970s, that mirrors what I’ve found in local precipitation records.
And master gardener Danna Fast has been recording the blooming dates of wildflowers near her home since 1982. Her white water lilies now open about two weeks earlier in summer than they did during the early 1980s, perhaps because the water they float in is getting warmer; June temperatures have indeed risen significantly here during the last three decades.
As we look ahead to how future climate changes might affect our landscapes and living things, we have less aid from computer models than we do for climate itself. Living things are not as easy to simulate as air masses are, but some investigators get a lot of press for trying anyway.
Such is the case for the flameout of sugar maples described in Bill McKibben’s Adirondack Life article. Bill got the story indirectly from other folks, who apparently got it from a study conducted by the USDA Forest Service. Their website displays colorful maps of expected changes in the optimal temperature ranges of various tree species, which, in turn, refer to a study that was published in 1998 in Ecological Monographs.
At first glance, the maps seem to warn of an imminent die-off. The tinted blobs representing sugar maples shrink away toward Canada, while oaks and hickories creep in to replace them. Maybe the Adirondacks really are slated to become botanical equivalents of the Blue Ridge and Smokies by 2100 AD.
But there’s a prominent button on the home page and readers are urged to click on it. When I clicked recently, I read the following: “In order to avoid the misinterpretation of our atlas, we want everyone to read the following section before making sense of the maps. We would like to stress here that our model is not predicting migration of species x—but rather the movement of suitable habitat for that species.”
In other words, these maps show preferred climatic conditions, not species migrations. Trees don’t simply uproot and chase after warmer temperatures. They also need suitable rainfall and soil types, effective ways to spread their seeds over long distances, and sufficient elbow room among established trees that can live for hundreds of years. One review of the subject in Frontiers in Ecology and the Environment pointed out that this kind of model “ig
nores transitional periods in which species may persist outside their normal climatic range,” and other studies have shown that human presence has altered the distributions of many species so much that it’s now impossible even to know exactly what their true natural ranges are in the first place. At this point I knew that I needed to speak with someone who knows maples much better than I do.
I turned to Mike Farrell, a forest ecologist who directs Cornell University’s maple research station near the ski jumps on the outskirts of Lake Placid. Mike has heard the stories about maples dying from global warming, but he doesn’t buy most of them. “Nobody really knows what’s going to happen here in the future,” he began, “but as far as oaks and hickories replacing maples in fifty or a hundred years is concerned, I think it’s unlikely. Sugar maples can actually grow better in warmer settings than the Adirondacks, and there’s a perfectly viable sugaring industry down in West Virginia. The main problem is with regeneration of new seedlings.” Acid rain, diseases, and deer-browsing already hurt the existing trees here and they also tend to keep new maples from replenishing the woodland ranks. According to Mike, those same problems might keep invasive oak and hickory seedlings in check, too.